HELGOLANDER MEERESUNTERSUCHUNGEN Helgol~inder Meeresunters. 35, 153-214 (1982) Phytogeographic distribution groups of benthic marine algae in the North Atlantic Ocean. A review of experimental evidence from life history studies C. van den Hoek Department of Systematic Botany, Biological Centre, The University; P.O. Box 14, 9750 AA Haren, Groningen, The Netherlands ABSTRACT: Experimentally determined lethal temperatures and temperatures limiting growth or reproduction in the life histories of 15 benthic algal species were used to infer possible phytogeo- graphic boundaries in the North Atlantic Ocean. These appeared to correspond closely with phytogeographic boundaries based on distribution data. Many boundaries appeared to be of a composite nature. For instance, the southernb oundary of Nemalion helminthoides is interpreted as a "southern reproduction boundary" on the N. Atlantic E. shore and a "southern lethal boundary" on the N. Atlantic W. shore. The northern boundary on both sides of the ocean is a "northern reproduction boundary". N. helminthoides is a typical representative of the "amphiaflantict emper- ate distribution group", to which seven other of the fifteen investigated species belong (Chondrus crispus, Desmarestia aculeata, .D viridis, Monostroma grevillei, Acrosiphonia "arcta'" with a comparable composite southern boundary; Rhodochorton purpureum with a "southern lethal boundary"). Polysiphonia ferulacea and Dictyota dichotoma are treated as representatives of the "amphiatlantic tropical-to-warm-temperate distribution group", and .P denudata as representative of the "amphiatlantic tropical-to-temperate group". .P harveyi belongs to the N.E. American temperate group and is bounded by a "northern reproduction boundary" and a "southern reproduc- tion boundary". This is one of the veryf ew species endemic to N.E. America. This poor endemism is ascribed to the vast adverse sediment shores and their additional acting as barriers to glacialn orth- south displacements of the flora; it is not related to the wide annual temperature fluctuations (> 20 °C) typical for N.E. America. The temperate algal flora of Japan, however, which is extremely rich in endemic species is subject to equally wide annual temperature fluctuations. Bonnemaisonia hamifera is such a Japanese endemic, which has been accidentally introduced into the North Atlantic Ocean where its life history seems to be disrupted: it is maintained mainly by vegetative propagation of the heteromorphic tetrasporophyte. The species of the "warm temperate Mediterra- nean-Atlantic group" are probably too stenothermous for life on N.E. American shores; they need annual temperature fluctuations < 20°C. Acrosymphyton purpuriferum seems to belong to this group, but arguments are presented to unite this species with A. caribaeum and to range it under the "amphiatlantic tropical-to-warm-temperate group". Clathromorphum circumscriptum belongs to the "Arctic distribution group" and has a "southern reproduction boundary" across the ocean along the 3 °C February isotherm. This species is able to survive temperatures of about 20 °C. Five amphiequatorial temperate species discussed in this paper and four in another related paper have similar maximum winter temperatures of 14-17 °C (mean monthly values) allowing reproduction. Their amphiequatorial distribution can be explained by assuming similar low temperatures in the euphotic zone along E. Pacific and E. Atlantic equatorial coasts i.e. in narrow inshore belts of intensified upwelling during the presumably intensified glacial circulation of the ocean gyres. © Biologische Anstalt Helgoland 0174-3597/82/0035/0153/$ 02.00 154 C. van den Hoek INTRODUCTION A recently finished revision {van den Hoek, 1981) and an earlier published one (van den Hoek, 1963) of the genus C1adophora (Chlorophyceae} have offered an opportunity to unravel the hitherto confused distribution patterns of 42 species of this genus in the North Atlantic Ocean {van den Hoek, 1979). These 42 species could be assigned to eight phytogeographic distribution groups which apparently had a general applicability to the marine algal flora of the North Atlantic Ocean (van den Hoek, 1982), and hypotheses were formulated about the nature of the phytogeographic boundaries of these distribu- tion groups. In principle, it is possible to test hypotheses about the nature of phytogeographic boundaries of individual benthic algal species by means of experimental studies on the environmental regulation of their life histories. A number of recent studies on the environmental regulation of life histories were, in general, not devised to answer questions on phytogeographic problems. Consequently the information contained in them is mostly too incomplete to solve such problems. Nonetheless, they contain a wealth of relevant information. Part of this information has been discussed elsewhere (van den Hoek, 1981}, and part of it will be reviewed in the present paper. BASIC GEOGRAPHIC DISTRIBUTION GROUPS IN THE NORTH ATLANTIC OCEAN The above mentioned eight distribution groups will be described below, together with two additional ones, }1( The amphiatlantic tropical-to-warm temperate group w i t h a n o r t h e a s t e r n e x t e n s i o n (Fig, .)1 Species of this group are entirely or Pig, .1 The distribution of Cladophora prolifera ,))I(( CI. coelothrix )O( and Hypnea musciformis )®( in the North Atlantic Ocean. ,1 (hatched): the amphiatlantic tropical-to-warm temperate group with a northeastern extension (CI, coelothrix type). 2 (stippled): the amphiatlantic tropical-to-warm temp- erate group with a northwestern extension (Hypnea musciformis type). 8 W° = 8 °C winter (Febru- ary) isotherm. 91 S° = 91 °C summer (August} isotherm Phytogeographic distribution 155 nearly restricted to the tropics in America and extend more or less far into the temperate zone in Europe. Cladophora coelothrix and .IC prolifera are examples. Their northern boundary is near the 8 °C February isotherm. This suggests that both species are limited to the north by lethal winter temperatures of the surface water. At least 50 species (on a total of c. 580 considered in a phytogeographic study of the N. Atlantic Ocean, see van den Hoek, 1979) seem to belong to this distribution group on the basis of distribution data. Experimental evidence supports the hypothesis that the northern boundary of one species of this group, Gracilaria foliifera, does indeed follow the 8 °C February isotherm, which can be considered as a "lethal boundary" (McLachlan & Edelstein, 1977; van den Hoek, 1982). In another species of this group, Centroceras clavulatum, the northern boundary was set by a combination of a summer isotherm in Europe (determined by the lowest temperature allowing growth) and a winter isotherm in America (determined by the lethal winter temperature) (van den Hoek, 1982). (2) The amphiatlantic tropical-to-warm temperate group w i t h a n o r t h w e s t e r n e x t e n s i o n (Fig. 1). The sole example, Hypnea muscifor- mis, has a clearly more northern distribution in North America than in Europe. This is probably caused by the fact that H. musciformis needs a high minimum summer temperature for its growth, but can at the same time survive very low winter tempera- tures (down to c. -1 °C). (3) The amphiatlantic tropical-to-temperate group (Fig. 2).The northern limit of the example, Cladophora vagabunda, follows approximately the 12 °C summer isotherm, which suggests that this limit is determined by the minimum summer temperature allowing sufficient growth and reproduction. On the basis of distribution data, at least fifteen species (on a total of c. 580 considered in a phytogeographic study of the N. Atlantic Ocean; see van den Hoek, 1979) can probably be attributed to this group, with northern boundaries along the 15°C to 12°C summer isotherms. Sphacelaria rigidula Kiitz. (= S. furcigera Kiitz.) has the same distribution as .1C vagabunda. Its northern boundary, the 12°C summer isotherm, is determined by the lowest summer Fig. 2. The distribution of vagabunda Cladophora ©( and stippled area) in the North Atlantic Ocean: the amphiatlantic tropical-to-temperate group (Cladophora vagabunda type) 156 C. van den Hoek temperature allowing formation of propagules (van den Hoek, 1982). This boundary is valid on a world wide scale. (4, 5) The amphiatlantic temperate group; with a southern boundary near a summer isotherm (4) and near a winter i s o t h e r m 5 ( ) (Fig. 3). Both Cladophora rupestris and .1C albida have northern boundaries approaching August isotherms (the 5 °C and 12 °C isotherms, respectively). This suggests that they are limited to the north by minimum summer temperatures allowing sufficient growth and/or reproduction. The southern boundary, however, of .1C rupestris is near the 20 °C summer isotherm, and of CI. albida near the 14 °C winter isotherm. This suggests that CI. rupestris is limited to the south by a lethal summer temperature; and CI. albida by a winter temperature which is sufficiently low to permit growth and/or reproduction. At its southern boundary in Texas .1C albida is a winter species, at its northern boundary a summer species. At least 40 species (on a total of c. 580 considered in a phytogeographic study of the N. Atlantic Ocean, see van den Hoek, 1979) can be attributed to group 4 on the basis of distribution data (van den Hoek, 1979, 1982). Their southern boundaries seem to follow the 17 °C-23 °C summer isotherms. The northern boundary of Callithamnion hookeri follows the 10 °C summer isotherm, and this corresponds with a minimum temperature of c. 10°C allowing sufficient growth. The southern boundary at the 25 °C summer isotherm corresponds with a lethal temperature of c. 27°C (Edwards, 1979; van den Hoek, 1982). In a comparable way, Dumontia contorta ----( D. incrassata] and Laminaria saccharina have southern boundaries along the 19 °C summer isotherm corresponding with somewhat higher lethal temperatures (van den Hoek, 1982; Rietema & Klein, 1981; Rietema, in prep.; Liining, 1980). The southern boundary of L. digitata, however, follows approximately the 10°C winter isotherm in Europe, which corresponds with the highest winter temperature allowing sufficient maturation of the female gametophyte. In America the southern boundary is at the 19 °C summer isotherm and this corresponds with a lethal temperature of c. 21 °C (van den Fig. .3 The distribution of .1C rupestris h(( stippled area: the amphiatlantic temperate group with a southern boundary near a summer isotherm). The distribution of CI. albida ,©( hatched area: the amphiatlantic temperate group with a southern boundary near a winter isotherm) Phytogeographic distribution 157 Hoek, 1982; Lfining, 1980). A comparatively small number of at least twelve benthic algal species (out of c. 580 in the N. Atlantic Ocean considered in a phytogeographic study) can be assigned to the distribution group 5. Scytosiphon lomentaria and Petalonia fascia are both limited to the south by the 17 °C winter isotherm. This corresponds with the highest temperature of c. 20 °C which allows, in winter, the initiation of erect thalli from crustose thalli in southern ecotypes (Lfining, 1981a, 1981b; van den Hoek, 1982). (6) The tropical Western Atlantic group (Fig. 4). Cladophoracatenata is an example. So far, no experimental data are available to explain the temperature responses of the species in this group, which embraces at least 250 species of the 450 inhabiting the tropical Western Atlantic region (and of the c. 580 species in the N. Atlantic Ocean considered in a phytogeographic study, cf. van den Hoek, 1979) which were considered in the context of phytogeographic comparisons (van den I-Ioek, 1975). (7) The American tropical-to-temperate group (Fig. 4). Only five species can be presently assigned to this group with some certainty. Gracilaria tikvahiae is limited to the North (in shallow embayments of the southern Gulf of St. Lawrence) by summer temperatures above 20° C allowing sufficient growth, and it is at the same time capable of surviving --2 °C below winter ice cover (Edelstein et al., 1976; Bird et al., 1977a, b, 1978; McLachlan and Edelstein, 1977; Edelstein, 1977; von den Hoek, 1982). Three of the four other species (e.g., Grinnellia americana) prefer similar shallow embayments, where they profit by high summer temperatures of up to 30 °C, but can survive freezing (van den Hoek, 1982). (8) The N.E. American temperate group (Fig. 4). Very fewspecies, for instance Chondria baileyana are endemic to the N.E. American shores between 30° N and 45 °N. Just like most species of the previous group, C. baileyana inhabits inshore water and embayments with wide annual temperature fluctuations. This scarcity of Fig. 4. The distribution of .1C catenata ,O( 6: the tropical Western Atlantic group); airalicarG tikvahiae ,®( 7: the American tropical-to-temperate group); bafleyana Chondria (e 8: the American temperate group); pellucida Cladophora ,®( 9: the warm temperate Mediterranean-Atlantic group); Sphacelaria (e, arctica :01 the Arctic group) 158 C. van den Hoek endemic species in the American temperate distribution group contrasts sharply with the large number of endemic species in the algal flora of the Japanese coasts which occupy a comparable latitudinal belt between 30° N and 45° N with equally wide annual tempera- ture fluctuations. Here 125 out of 170 "indicator species" (Funahashi, 1974} are endemic. The Japanese flora comprises in total c. 750 Chloro-, Rhodo-, and Phaeophyceae. Experimental evidence indicates that Campylaeophora hypneoides J. Ag. (Notoya, 1979) is limited to the north by the 15°C summer isotherm, which indicates the minimum temperature necessary for sufficient growth; and to the south by the 27° C summer isotherm, which corresponds approximately with a lethal temperature of 30 °C. As this species is able to survive in its northern range --2 °C, it can bridge a temperature range of --2 °C to 30 °C. Sargassum muticum, a Japanese species which has been accidentally introduced in N.W. America und N,W. Europe, has a comparable wide temperature range. It is likely that the poverty of the endemic flora along N.E. American coasts between 30 °N and 45° N can be attributed to the adverse nature of the sediment coasts to the growth of benthic macroalgae. Moreover, these adverse coasts have probably acted as barriers to north-south and south-north displacements of the benthic algal flora during glaciatibns and have thus caused its impoverishment and prevented successful specia- tion. (9) The warm temperate Mediterranean-Atlantic group (Fig. 4). Cladophora pellucida is one of the numerous examples. At least 180 species of the 430 species inhabiting European Atlantic shores and considered in a phytogeographic comparison embracing c. 580 species in the N. Atlantic Ocean, can be attributed to this distribution group (van den Hoek, 1975, 1979). In one species, Saccorhlza polyschides (see Norton, 1977, for the regulation of its life history}, the southern boundary in N.W. Africa is situated at about the 15°C February isotherm. This corresponds with the maximum winter temperature of 17 °C allowing maturation of female gametophytes (at still higher temperatures they only grow vegetatively}. The winter temperature lethal to female gametophytes and young sporophytes is 3 °C. The summer temperature lethal to sporophytes of S. polyschldes is 25 °C. The corresponding lethal boundaries are the 4 °C winter isotherm in the north and the 23 °C summer isotherm in the south, embracing 19°C of mean monthly temperatures of the surface water. This maximum temperature span makes S. polyschides unfit for inhabiting temperate American coasts where all species must be capable of bridging temperature spans of more than 20°C (mean monthly temperature}. This is a very obvious cause of the absence of species endemic to European coasts from N.E. American coasts (van den Hoek, 1982}. .;5 polyschldes bridges, along African and European shores, an enormous latitudinal span of 25 °N to 65 °N. Nonetheless it is incapable of inhabiting the N.F.. American coasts as a result of its temperature responses. Endemic species with still narrower latitudinal spans in Europe are probably even less fit for life on American shores than Saccorhiza polyschldes. (10) The Arctic group {Fig. 4). Sphacelarla arctica is an example which follows the general trend of Arctic species to penetrate more or less far into the cool temperate zone. About at least 20 species (on a total of 580 considered in a phytogeo- graphic study of the N. Atlantic Ocean, see von den Hoek, 1979} can be assigned to the Arctic group on the basis of distribution data (van den Hoek, 1979}. The southern boundary of Saccorhiza dermatodea is near the 4 ° February isotherm in Europe, and the 15°C August isotherm in America. The 4 °C February isotherm corresponds with the Phytogeographic distribution 159 highest winter temperature (5°C) allowing sufficient maturation of the female gametophyte. The 15 °C summer isotherm in America corresponds with a lethal tempera- ture of 17°C for the sporophyte (Norton, 1977; van den Hoek, 1982). Sphacelaria arctica has also a southern boundary set, in Europe, by the maximum temperature in winter allowing reproduction (the 4 °C winter isotherm); and, in America, by a lethal summer temperature (the 16 °C August isotherm). These boundaries allow S. arctica to inhabit the Baltic Sea with winter temperatures lower than 4 ° and summer temperatures lower than 16 °C in the greatest part. Apparently S. arctica can tolerate salinities as low as 05 % (van den Hoek, 1982). NEW EVIDENCE FROM LIFE HISTORY STUDIES IN SUPPORT OF THE BASIC GEOGRAPHIC DISTRIBUTION GROUPS When one tries to extract phytogeographically relevant information from studies on life histories, one should realize that the species may be confronted with at least the following boundaries corresponding with various types of limiting temperature condi- tions which should all be checked when perusing the results of the above studies. In essence, the significance of these boundaries has been previously stressed by Hutchins (1947): (1) The "northern lethal boundary" corresponds with the lowest winter temperature which a species can survive for a period of c. 1/2-1 month. (2) The "northern growth boundary" corresponds with the lowest summer temperature which, in a period of several months, allows sufficient growth. )3( The "northern reproduction boundary" corresponds with the lowest summer temperature which in a period of several months allows sufficient reproduction. Often "sufficient growth" ends with reproductive matura- tion but in many cases the formation of reproductive structures is governed by other temperatures than those governing growth. (4) The "southern lethal boundary" corres- ponds with the highest summer temperature which a species can survive during a period of 1/2-1 month. (5) The "southern growth boundary" corresponds with the highest winter temperature which, in a period of several months, allows sufficient growth. (6) The "southern reproduction boundary" corresponds with the highest winter temperature which, in a period of several months, allows sufficient reproduction. For species with an alternation of generations the six boundaries described above should be determined in principle for both generations. However, in species with an isomorphic alternation both generations have mostly, but not always, similar tempera- ture responses. Photoperiodic responses may modify the temperature responses and should consequently always be taken into account. It is, moreover, necessary to make an estimate of the difference between the experimental temperature limits and the corresponding mean monthly temperatures of the sea surface in order to be able to "translate" an experimental temperature limit to a boundary approaching a surface isotherm. These estimates can be made in the following way. A "southern lethal summer temperature" is considered to correspond with the maximum August temperature of the uppermost surface water, which is taken, in its turn, to correspond approximately with a three centigrades lower mean August tempera- 160 C. van den Hoek ture (cf. Earle, 1969). It should be realized, however, that this difference can be greater in protected lagoons. A "southern growth and/or reproduction winter temperature" corresponds with the mean maximum February temperature (for the correct temperature has to be provided each year); which is taken, in its turn, to correspond with a two centigrades lower mean February temperature for temperatures above 10 °C, and with a one centigrade lower mean February temperature below 10 °C. A "northern lethal winter temperature" is considered to correspond with the minimum February temperature of the surface layer. This corresponds with a one centigrade higher mean February temperature (two centigrades for temperatures above 10 oc). A "northern growth and/or reproduction temperature" is considered to correspond with a mean minimum August temperature of surface water, and this with a two centigrades higher mean August temperature. The present study presents evidence for the validity of the following phytogeo- graphic distribution groups: the amphiatlantic tropical-to-warm temperate group with a northeastern extension (group 1, Fig. 1); the amphiatlantic tropical-to-temperate group (group 3, Fig. 2); the amphiatlantic temperate group of the Cladophora rupestris type (group 4, Fig. 3) and a type with features of the Cl. rupestris as well as the Cl. albida type (Fig. 3) (groups 4 and 5); the N.E. American temperate group (group 8, Fig. 4); the Japanese equivalent of the American temperate group (group 8); and the Arctic group (group 10, Fig. 4). Table 1 lists the species and corresponding distribution groups for which the evidence is reviewed in the present paper. Distribution data from the following publications were used for the construction of the distribution maps; Abbott & Hollenberg (1976); Adey (1973); Adey et al. (1976); Ardr~ (1970, 1971); Baardseth (1941); Boudouresque & Perret (1977); Brauner (1975); Breeman (1979); Chamberlain (1965); Chapman (1969); Chiang (1972, 1973); Chihara & Yoshizaki (1970); Christensen & Thomsen (1974); Dawes (1974); Dawson (1946); Day (1969); De May et al. (1977); De S~ve et al. (1979); Diannelidis et al. (1977); Dixon & Irvine (1977); Edelstein (1964); Edelstein et al. (1969, 1970); Edelstein & McLachlan (1966, 1968); Edwards (1970a, 1979); Earle (1969); Feldmann (1942, 1954); Feldmann- Mazoyer (1940); Funahashi (1966, 1967, 1974); Funk (1955); Furnari et al. (1977); Furnari & Scammacca (1973); Giaccone (1969, 1978); Giaccone & Longo (1976); Hamel (1930); Hamel (1931-39); Haritonidis & Tsekos (1974); Den Hartog (1959); Hawkes et al. (1978); Hirose & Kajimura (1973); van den Hoek & Donze (1967); Humm (1979); Jaasund (1965); John et al. (1977); Jonsson (1912); Juett et al. (1976); Kapraun (1977); Knaggs (1967); Konno (1973); Kornmann & Sahling (1977); Kiihnemann (1972); Lawson (1966); Lawson et al. (1975); Lawson & John (1977); Lee (1980); Lee & Kim (1977); Lee & Lee (1976); Lindauer et al. (1961); Lipkin (1972); Levring (1960, 1974); Makienko (1975); Mathieson (1979); Moura Baptista (1977); Munda (1977, 1978, 1980); Nakaniwa (1975); Newton (1931); Nizamuddin et al. (1979); Ogawa & Machida (1976); Oliveira Filho (1977); Orris & Taylor (1973); Pankow (1971); PapenfuB (1964); Pedersen (1976); Perez-Cirera (1975); Pielou (unpublished list of distribution records along American coasts); Price et al. (1978); Rhodes (1970); Rueness (1977); Santelices (1980); Scagel (1957); S~apova & Voz~inskaja (1960); Schmidt (1931); Schneider (1976); Schneider et al. (1979); Searles & ¢0 o reviewed is categories Dictyotales Ceramiales Ceramiales Nemaliales Nemaliales Gigartinales Desmarestiales Ulotrichales Acrosiphoniales Dictyosiphonales Ceramiales Bonnemaisoniales Cryptonemiales Cryptonemiales limits Taxonomic Phaeophyceae, Rhodophyceae, Rhodophyceae, Rhodophyceae, Rhodophyceae, Rhodophyceae, Phaeophyceae, Phaeophyceae, Desmarestiales Chlorophyceae, Chlorophyceae, Phaeophyceae, Rhodophyceae, Rhodophyceae, Rhodophyceae, Rhodophyceae, phytogeographic Harv. Batt, of ex Sj6st. nature Grev. Rosenv. in With.) Lamour. Ag. Ag.) (Strbmfeldt) on the treated Lamour. (Lightf.) (Vell. Lamour. Mfill.) Wittr. J. Hariot (J. Suhr (Dillw,) (L.) F. (Thur.) (Ditlw.) Bailey evidence species (Huds.) ferulacea denudata purpureum Stackh. aculeata viridis (O. greviltei "arcta'" Grey. harveyi hamifera purpuriferum circumscriptum which dichotoma helminthoides crispus latifolia for Representative Hook, Dictyota Polysiphonia Polysiphonia in Rhodochorton Nemalion Chondrus Desmarestia Desmarestia h4onostroma Acrosiphonia Punctaria Polyslphonia Bonnemaisonia Acrosymphyton Clathromorphum Foslie grofl~s distribution southern southern a a group with isotherm isotherm Table 1. Species and corresponding Name of distribution Amphiatlantic Amphiatlantic tropical-to-warm-temperate with a northeastern extension Amphiatlantic tropical-to-warm-temperate with a northwestern extension Amphiatlantic tropical-to-temperate Amphiatlantic temperate: boundary near a summer Amphiatlantic temperate: with boundary near a winter America Tropical Western Atlantic American tropical-to-temperate N.E. American temperate Japanese temperate Europe/North Africa Warm temperate Mediterranean-Atlantic Amphiatlantic, polar Arctic Group No. (A) (1) (2} (3) (4} (5) (B} (6) (7) (8) (8') (C) (9) (D) (10} 162 C. van den Hoek Schneider (1978); Sears (1971); Setchell & Gardner (1919-20, 1925); South (1976}; South & Hooper (1980); Stephenson & Stephenson (1972); Tanaka & Itono (1972); Taylor (1945, 1957, 1960); Tokida (1954); Tsekos & Haritonidis (1977); Tsuda & Wray (1977); Velas- quez et al. (1975); Velasquez & Lewmanomont (1975); Voz~inskaja (1964, 1965); Wilce (1959); Womersley (1965, 1967); Womersley & Bailey (1970); Zaneveld (1972, 1976); Zimmermann & Livingston (1976); Zinova (1953, 1955, 1967). The amphiatlantic tropical-to-warm temperate group with a northeastern extension Dictyota dlchotoma Life history Dictyota dichotoma has an isomorphic alternation between a tetrasporophyte gener- ation and a dioecious gametophyte generation with oogamy. Dlctyota dichotoma grows rapidly, and several generations may follow each other in the course of one year. Sporophytes and gametophytes are present at the same time and not connected in a seasonal sequence. The production of eggs is claimed to have a lunar periodicity (Miiller, 1962~ Vielhaben, 1963}. The influence of temperature on reproduction has not been investigated. Near its northern boundary in N.E. America (in N. Carolina) .91 dichotoma disap- pears from January through March. It appears to survive the winter in the form of small decumbent germlings which resume growth by the formation of erect blades after transfer to higher temperatures (Richardson, 1979). These gerrnlings remain "dormant" under "winter conditions" (12 °C) in culture, and resume growth (by formation of blades) after transfer to 24 °C in culture. At Beaufort (N. Carolina) inshore populations of D. dichotoma must be able to survive intermittently winter temperatures as low as 3 °C (Brauner, 1975). In Virginia this lowest temperature may even be 2 °C (Humm, 1979). This suggests a lethal winter temperature near 2 °C. In short term experiments (12 h exposures) Dictyota dichotoma from Roscoff (Brittany, W. France} survived 3 °C, but not --2 °C (Biebl, 1959). However, in the same type of experiments conducted with material from Puerto Rico, D. dichotoma survived 5 °C, but not 3 °C (Biebl, 1962}. This indicates that D. dlchotoma embraces a number of temperature ecotypes. There are comparable differences in the highest temperatures tolerated: plants from Roscoff survived 27 °C, but not 30 °C; and plants from Puerto Rico survived 32 °C, but not 35 °C. At 12 °C Dictyota germlings did not grow (Richardson, 1979}. Apparently the minimum temperature allowing growth must be higher than 12 °C. Geographic distribution in the North Atlantic Ocean The northern boundary of .91 dichotoma in N.E. America approximates the 2 °C February isotherm, which limit can be explained as a "northern lethal boundary", approximately corresponding with a minimum February temperature of 1 °C. The northern boundary in Europe approximates the 13 °C summer isotherm, which limit can be explained as a "northern growth and reproduction boundary". This would